June 2020
Volume 61, Issue 7
ARVO Annual Meeting Abstract  |   June 2020
Predictive coding of motion in the primate retina
Author Affiliations & Notes
  • Michael B Manookin
    Ophthalmology, University of Washington, Seattle, Washington, United States
  • Footnotes
    Commercial Relationships   Michael Manookin, None
  • Footnotes
    Support  NIH Grant R01-EY027323, NIH Grant R01-EY029247, NIH Grant P51 OD-010425, Research to Prevent Blindness Unrestricted Grant, NIH Grant P30-EY001730, Alcon Young Investigator Grant
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 5138. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Michael B Manookin; Predictive coding of motion in the primate retina. Invest. Ophthalmol. Vis. Sci. 2020;61(7):5138.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Purpose : Many animals use information about visual motion to guide their behavior. Motion produces strongly stereotyped patterns of light stimulation on the retina, and these patterns (correlations) comprise a potentially rich source of information that can be utilized in behaviorally salient tasks such as predicting the future position of a moving object. Such prediction could compensate for processing delays inherent in phototransduction. Indeed, neurons have been found in several vertebrate retinas that utilize motion correlations in this way (Palmer et al 2015 PNAS; Berry et al. 1999 Nature; Schwartz et al 2007 Neuron). Whether similar properties are found in the primate retina is not known. Thus, we sought insight into this question by studying four ganglion cell types that provide input to motion processing pathways in primates.

Methods : We recorded the responses of ganglion cells to different motion correlations. Five different classes of motion correlation were tested and compared to stimuli lacking motion correlations. Spike outputs and excitatory synaptic inputs were measured to these stimuli in On- and Off-type parasol and smooth monostratified ganglion cells of the macaque monkey retina. The information about each stimulus class contained in the neural responses was determined by calculating the time-shifted mutual information.

Results : The spike outputs of parasol and smooth monostratified ganglion cells showed higher information rates for motion stimuli relative to a stimulus lacking motion correlations. Further, one type of motion correlation caused these cells to encode information about the future state of the stimulus. This predictive coding was present both in the spike output of the ganglion cells and in their excitatory synaptic inputs from diffuse bipolar cells. Further, this behavior could not be explained using canonical receptive-field models.

Conclusions : The processing of light stimuli by the visual system involves unavoidable delays. These delays are particularly problematic for neural circuits tasked with encoding motion in the environment. Here we identify four ganglion cell pathways in the primate retina that compensate for these delays by utilizing the inherent spatiotemporal correlations in visual motion to predict future motion. This work builds on previous findings that identify the parallel pathways that give rise to motion processing and neural mechanisms mediating motion sensitivity in the primate retina.

This is a 2020 ARVO Annual Meeting abstract.


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.